trial

app.py

@@ -1,13 +1,8 @@
 import json
 from typing import Dict
-
-from datasets import load_dataset
-
-from db.schema import Feedback, Response
-from db.crud import ingest, read, save_feedback
-import pandas as pd
+from utils.loaders import load_data
+from db.crud import read
 import streamlit as st
-from datetime import datetime
 import os
 from dotenv import load_dotenv
 from views.intro_screen import welcome_screen
@@ -18,6 +13,7 @@ from css.layout import custom_css
 load_dotenv()
 VALIDATION_CODE = os.getenv("VALIDATION_CODE")
 
+
 def initialization():
     """Initialize session state variables."""
     if "current_index" not in st.session_state:
@@ -58,21 +54,6 @@ def reset_survey():
     st.session_state.start_new_survey = True
 
 
-def load_data():
-    try:
-        data = pd.read_csv("data/gemini_results_subset.csv")[:5]
-        return data
-    except Exception as e:
-        repo_name = os.getenv("DATA_REPO")
-        data_files = os.getenv("LLAMA_DATA_FILES")
-        print("data_files", data_files)
-        HF_TOKEN = os.getenv("HF_TOKEN")
-        dataset = load_dataset(repo_name, token=True, data_files=data_files, revision="main")
-        dataset.set_format(type='pandas')  ## converting it into pandas
-        df = dataset["train"][:]
-        return df[:5]
-
-
 def ui():
     """Main function to control the survey flow."""
     custom_css()

utils/loaders.py

@@ -0,0 +1,25 @@
+import pandas as pd
+import os
+
+from datasets import load_dataset
+from dotenv import load_dotenv
+
+load_dotenv()
+HF_TOKEN = os.getenv("HF_TOKEN")
+REPO_NAME = os.getenv("DATA_REPO")
+DATA_FILES = os.getenv("LLAMA_DATA_FILES")
+
+
+def load_data():
+    try:
+        #TODO: change this to load the data from the database (buggy for debugging)
+        data = pd.read_csv("../data/gemini_results_subset.csv")[:5]
+        return data
+    except Exception as e:
+
+        print("data not found, loading from huggingface dataset")
+
+        dataset = load_dataset(REPO_NAME, token=True, data_files=DATA_FILES, revision="main")
+        dataset.set_format(type='pandas')  ## converting it into pandas
+        df = dataset["train"][:]
+        return df[:5]

views/nav_buttons.py

@@ -38,7 +38,7 @@ def navigation_buttons(data, ratings_v, ratings_p0, ratings_p1):
 
     col1, col2, col3 = st.columns([1, 1, 2])
 
-    with col1:  # Back button
+    with col1:  # Back button #TODO fix
         if st.button("Back") and current_index > 0:
             st.session_state.current_index -= 1
             st.rerun()

views/questions_screen.py

@@ -1,7 +1,4 @@
-import json
-from typing import Dict
-
-from db.schema import Feedback, Response
+from db.schema import Response
 import streamlit as st
 from datetime import datetime
 import os
@@ -9,7 +6,17 @@ from dotenv import load_dotenv
 from views.nav_buttons import navigation_buttons
 
 load_dotenv()
-VALIDATION_CODE = os.getenv("VALIDATION_CODE")
+if "VALIDATION_CODE" in os.environ:
+    VALIDATION_CODE = os.getenv("VALIDATION_CODE")
+if "DATA_REPO" in os.environ:
+    REPO_NAME = os.getenv("DATA_REPO")
+else:
+    print("DATA_REPO not found in environment variables.")
+if "LLAMA_DATA_FILES" in os.environ:
+    DATA_FILES = os.getenv("LLAMA_DATA_FILES")
+else:
+    print("LLAMA_DATA_FILES not found in environment variables.")
+
 
 
 def survey_completed():
@@ -30,7 +37,7 @@ def render_query_ratings(query_label, config, query_key, current_index, has_pers
     """Helper function to render ratings for a given query."""
     st.markdown(f"### {query_label}")
     st.write(config[query_key])
-    columns = st.columns(3 if has_persona_alignment else 2)
+    # columns = st.columns(3 if has_persona_alignment else 2)
     columns = st.columns(3)
     options = [0, 1, 2, 3, 4]
 
@@ -47,16 +54,13 @@ def render_query_ratings(query_label, config, query_key, current_index, has_pers
         relevance_rating = st.radio("Relevance:",
                                     options, key=f"rating_{query_key}_relevance_{current_index}",
                                     format_func=lambda x:
-                                    ["N/A", "Not Relevant", "Somewhat Relevant", "Relevant", "Unclear"][x],)
+                                    ["N/A", "Not Relevant", "Somewhat Relevant", "Relevant", "Unclear"][x], )
     with columns[2]:
         clarity_rating = st.radio("Clarity:",
-                                  options = [0, 1, 2, 3],
+                                  options=[0, 1, 2, 3],
                                   key=f"rating_{query_key}_clarity_{current_index}",
                                   format_func=lambda x: ["N/A", "Not Clear", "Somewhat Clear", "Very Clear"][x])
 
-
-
-
     return {
         "clarity": clarity_rating,
         "relevance": relevance_rating,
@@ -117,4 +121,3 @@ def questions_screen(data):
         navigation_buttons(data, query_v_ratings["clarity"], query_p0_ratings["clarity"], query_p1_ratings["clarity"])
     except IndexError:
         print("Survey completed!")
-